A Perception-based Quality Metric for Omnidirectional Images

A key component in VR systems is omnidirectional content, which provides 360-degree views of scenes. At a given time, only a portion of a content, called viewport, is displayed using head-mounted displays. In this work, we develop an objective quality metric, called Weighted Viewport PSNR (W-VPSNR), for omnidirectional content of spatially-varying quality. Through experiment results, it is found that the W-VPSNR metric well correlates with the subjective quality scores.

[1]  Truong Cong Thang,et al.  A subjective study on QoE of 360 video for VR communication , 2017, 2017 IEEE 19th International Workshop on Multimedia Signal Processing (MMSP).

[2]  Anh T. Pham,et al.  A Study on Quality Metrics for 360 Video Communications , 2018, IEICE Trans. Inf. Syst..

[3]  Joseph C. Besharse,et al.  The retina and its disorders , 2011 .

[4]  Krista A. Ehinger,et al.  SUN database: Large-scale scene recognition from abbey to zoo , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[5]  Marios S. Pattichis,et al.  Foveated video quality assessment , 2002, IEEE Trans. Multim..

[6]  Anh T. Pham,et al.  An Optimal Tile-Based Approach for Viewport-Adaptive 360-Degree Video Streaming , 2019, IEEE Journal on Emerging and Selected Topics in Circuits and Systems.